Predictors of Subsidence and its Clinical Impact After Expandable Cage Insertion in Minimally Invasive Transforaminal Interbody Fusion

Abstract

Study Design. Retrospective review of prospectively collected multisurgeon data. Objective. Examine the rate, clinical impact, and predictors of subsidence after expandable minimally invasive transforaminal lumbar interbody fusion (MI-TLIF) cage. Summary of Background Data. Expandable cage technology has been adopted in MI-TLIF to reduce the risks and optimize outcomes. Although subsidence is of particular concern when using expandable technology as the force required to expand the cage can weaken the endplates, its rates, predictors, and outcomes lack evidence. Materials and Methods. Patients who underwent 1 or 2-level MI-TLIF using expandable cages for degenerative lumbar conditions and had a follow-up of >1 year were included. Preoperative and immediate, early, and late postoperative radiographs were reviewed. Subsidence was determined if the average anterior/posterior disc height decreased by >25% compared with the immediate postoperative value. Patient-reported outcomes were collected and analyzed for differences at the early (<6 mo) and late (>6 mo) time points. Fusion was assessed by 1-year postoperative computed tomography. Results. One hundred forty-eight patients were included (mean age, 61 yr, 86% 1-level, 14% 2-level). Twenty-two (14.9%) demonstrated subsidence. Although statistically not significant, patients with subsidence were older, had lower bone mineral density, and had higher body mass index and comorbidity burden. Operative time was significantly higher (P = 0.02) and implant width was lower (P < 0.01) for subsided patients. Visual analog scale-leg was significantly lower for subsided patients compared with nonsubsided patients at a >6 months time point. Long-term (>6 mo) patient-acceptable symptom state achievement rate was lower for subsided patients (53% vs. 77%), although statistically not significant (P = 0.065). No differences existed in complication, reoperation, or fusion rates. Conclusions. Of the patients, 14.9% experienced subsidence predicted by narrower implants. Although subsidence did not have a significant impact on most patient-reported outcome measures and complication, reoperation, or fusion rates, patients had lower visual analog scale-leg and patient-acceptable symptom state achievement rates at the >6-month time point. Level of Evidence. Level 4.